摘要
声速剖面反演基于历史数据获取声速剖面,是海洋环境声层析中的一种方法。传统声速剖面反演法很难获取大面积海域的非等效声速剖面,得出的结果一般都是等效声速剖面。为解决该问题,提出一种基于反向传播(Back Propagation,BP)神经网络的非等效声速剖面反演方法。对比结果表明,该方法能有效反演非等效声速剖面并且反演结果相比传统声速剖面反演具有更高的精度。
The inversion of sound velocity profiles based on historical data is a method in marine environmental acoustic tomography.Traditional sound velocity profile inversion methods are difficult to obtain non-equivalent sound velocity profiles in large areas of the sea,and the results obtained are generally equivalent sound velocity profiles.To address this issue,a non equivalent sound velocity profile inversion method based on Back Propagation(BP)neural network is proposed.The comparison results show that this method can effectively invert non equivalent sound velocity profiles and the inversion results have higher accuracy compared to traditional sound velocity profile inversion.
作者
于佩衡
YU Peiheng(Hangzhou Applied Acoustics Research Institute,Hangzhou 310000,China)
出处
《电声技术》
2024年第7期52-54,共3页
Audio Engineering
关键词
声速剖面反演
反向传播(BP)
非等效声速剖面
sound speed speed profile inversion
Back Propagation(BP)
non-equivalent sound velocity profile
